The wave group has recently published work in the Computers and Fluids journal. The work focuses on the performance analysis of the internally developed tsunami code: Volna-OP2. Volna-OP2 is a robust and efficient code capable of simulating the complete life cycle of a tsunami whilst harnessing the latest High Performance Computing architectures.
An in-depth error analysis of the numerical scheme implemented in the code is given, with a novel decomposition of the numerical errors into the dispersion and dissipation components explored. Most tsunami codes exhibit amplitude smearing and/or phase lagging/leading, so the error decomposition shown here is a new approach and novel tool for explaining these occurrences.
To date, Volna-OP2 has been widely used by the tsunami modelling community. In particular its computational efficiency has allowed various sensitivity analyses and uncertainty quantification studies. Due to the number of simulations required, there is always a trade-off between accuracy and runtime when carrying out these statistical studies. In this paper, the performance scalability of the GPU version of the code is assesed by simulating a realistic tsunami scenario based on a hypothetical submarine landslide at the Rockall Bank Slide Complex, which is located in the North East Atlantic. The analysis presented in this paper will guide future users towards an acceptable level of accuracy within a given runtime.
Please visit the following link to access the full paper: https://doi.org/10.1016/j.compfluid.2020.104649
Dan Giles
Irish Research Council Government of Ireland Postgraduate Scholar